The role of soluble 5'-nucleotidases in the conversion of nucleotide analogs: metabolic and therapeutic aspects.

A variety of anti-proliferative drugs is based on the structure of purine or pyrimidine nucleosides. These compounds, after phosphorylation, act as analogs of natural nucleotides. In vivo they are recognized by enzymes that transform them either into anti-metabolites targeted to the synthesis of DNA or RNA, or to inactive products of detoxification. 5'-Nucleotidases of different specificity and cellular localization can either remove the phosphate residue from the 5'- position of (deoxy) nucleotide or transfer it from nucleoside monophosphate onto other nucleosides. Such a nucleoside phosphotransferase activity also works with analogs of canonical nucleotides and nucleosides. The majority of nucleoside analogs is metabolized by intracellular cytoplasmic or mitochondrial 5'-nucleotidases and only few reactions proceed on the cell surface. This review summarizes our knowledge of cytoplasmic and mitochondrial forms of 5'-nucleotidases and focuses on their ability to dephosphorylate different analogs of canonical nucleoside 5'-monophosphates. The involvement of 5'-nucleotidases in the phosphotransfer reaction with some nucleoside analogs has been also presented. The importance of the reactions catalyzed by 5'-nucleotidases in clinical resistance to nucleoside-based drugs used in the treatment of cancer or viral diseases, as well as in activation of pro-drugs has been highlighted.

Novel approaches in the synthesis of batracylin and its analogs: rebirth of an old player?

Batracylin (8-aminoisoindol[1,2-b]-quinasolin-12(10H)-one, BAT), a heterocyclic amine, was isolated in 1978 (NCI, Bethesda, USA) in the course of search for the new anticancer drugs. It showed high in vitro and in vivo anticancer activities against murine leukemia P338 and colon adenocarcinoma 38. Mechanism of action of BAT is still not completely clear. It was reported, that BAT is a topoisomerase II inhibitor and induces unscheduled DNA synthesis (UDS) in non-proliferating cells. Low solubility of BAT in water, high toxicity and necessity of high drug dosing are major limitations of its use as a chemotherapeutic drug. As a result, new BAT analogs were synthesized to improve its pharmacological properties. The modifications of BAT chemical structure include various substituents introduced to isoindoloquinazoline moiety (Cl, Br, NO(2), CH(2), NH(2), Me, CO(2)Me, OMe). It has been shown that the desamino derivative and the 8-aza analog of BAT retained the ability to inhibit topoisomerase II but did not induce unscheduled DNA synthesis. While less active than BAT, these analogs were cytotoxic toward CCRF-CEM leukemia cells. The isoindolo [2,1-a]benzimidazole derivatives were inactive as topoisomerase II inhibitors and, in general, failed to exhibit comparable antitumor activity or to induce unscheduled DNA synthesis. Batracylin was acylated with aminoacids, dipeptides, tripeptides to increase its solubility in water. Other modifications include introduction of nitrogen atom to ring A or D, extension of polycyclic ring 4, reduction of ring B from six- to five-membered one, and obtaining of benzimidazole, indole or derivatives containing a fucose ring. A series of novel BAT analogs bearing sugar residues and thiocarbonyl aminoacids, which provided better solubility in water and high cytostatic activity have been designed. Also, new azabatracylines, where aniline ring was replaced by pyridine or other substituted quinazolines, have been obtained. This paper reviews the most important approaches in batracylin synthesis and its analogs and presents structure-reactivity relationships for these compounds.

Imidazoacridinone-dependent lysosomal photodestruction: a pharmacological Trojan horse approach to eradicate multidrug-resistant cancers.

Multidrug resistance (MDR) remains a primary hindrance to curative cancer therapy. Thus, introduction of novel strategies to overcome MDR is of paramount therapeutic significance. Sequestration of chemotherapeutics in lysosomes is an established mechanism of drug resistance. Here, we show that MDR cells display a marked increase in lysosome number. We further demonstrate that imidazoacridinones (IAs), which are cytotoxic fluorochromes, undergo a dramatic compartmentalization in lysosomes because of their hydrophobic weak base nature. We hence developed a novel photoactivation-based pharmacological Trojan horse approach to target and eradicate MDR cancer cells based on photo-rupture of IA-loaded lysosomes and tumor cell lysis via formation of reactive oxygen species. Illumination of IA-loaded cells resulted in lysosomal photodestruction and restoration of parental cell drug sensitivity. Lysosomal photodestruction of MDR cells overexpressing the key MDR efflux transporters ABCG2, ABCB1 or ABCC1 resulted in 10- to 52-fold lower IC(50) values of various IAs, thereby restoring parental cell sensitivity. Finally, in vivo application of this photodynamic therapy strategy after i.v. injection of IAs in human ovarian tumor xenografts in the chorioallantoic membrane model revealed selective destruction of tumors and their associated vasculature. These findings identify lysosomal sequestration of IAs as an Achilles heel of MDR cells that can be harnessed to eradicate MDR tumor cells via lysosomal photodestruction.

Evaluation of the acute toxicity of perfluorinated carboxylic acids using eukaryotic cell lines, bacteria and enzymatic assays.

The acute biological activity of a homologous series of perfluorinated carboxylic acids - perfluorohexanoic acid (PFHxA), perfluoroheptanoic acid (PFHpA), perfluorooctanoic acid (PFOA), perfluorononanoic acid (PFNA) and perfluorodecanoic acid (PFDA) - was studied. To analyze the potential risk of the perfluorinated acids to humans and the environment, different in vitro toxicity test systems were employed. The cytotoxicity of the chemicals towards two different types of mammalian cell lines and one marine bacteria was investigated. The viability of cells from the promyelocytic leukemia rat cell line (IPC-81) and the rat glioma cell line (C6) was assayed calorimetrically with WST-1 reagent. The evaluation was combined with the Vibrio fischeri acute bioluminescence inhibition assay. The biological activity of the compounds was also determined at the molecular level with acetylcholinesterase and glutathione reductase inhibition assays. This is the first report of the effects of perfluorinated acids on the activity of purified enzymes. The results show these compounds have a very low acute biological activity. The observed effective concentrations lie in the millimole range, which is well above probable intracellular concentrations. A relationship was found between the toxicity of the perfluorinated carboxylic acids and the perfluorocarbon chain length: in every test system applied, the longer the perfluorocarbon chain, the more toxic was the acid. The lowest effective concentrations were thus recorded for perfluorononanoic and perfluorodecanoic acids.

Ecto-5'-nucleotidase (eN, CD73) is coexpressed with metastasis promoting antigens in human melanoma cells.

Upregulated expression of eN has been found in the highly invasive human melanoma cell lines but neither in melanocytes nor in primary tumor cells. Membrane proteins associated with cell adhesion and metastasis: alpha5-, beta1-, beta3-integrins, and CD44 were elevated gradually in accordance with increasing metastatic potential. alphav-integrin was seen mostly in aggressive melanomas. The expression of eN correlated with a number of metastasis-related markers and thus may have a function in the process. eN activity went parallel with its amount in all cells. Concanavalin A strongly inhibited the enzyme in a noncompetitive way. Clustering of eN protein in overexpressing cells by ConA-treatment increased the enzyme association with the heavy cytoskeletal complexes. A similar shift towards cytoskeletal fractions took also place with other membrane proteins coexpressed with eN. This ConA-induced association may reflect a putative interaction of eN with physiological ligand, that upon interaction, aggregates protein components of lipid rafts and triggers signaling pathway that may be intrinsically involved in cell-stroma adhesion.

AMP deaminase in vitro inhibition by xenobiotics A potential molecular method for risk assessment of synthetic nitro- and polycyclic musks, imidazolium ionic liquids and N-glucopyranosyl ammonium salts.

The usefulness of in vitro AMP deaminase inhibition was examined as a potential molecular method in risk assessment of xenobiotics. The enzyme participates in the principal purine nucleotide interconversion and degradation pathways, and its absence caused perturbations in the cellular ATP pool. The compounds selected were synthetic musks with a known negative environmental impact and the toxicologically unknown ionic liquids and N-glucopyranosyl ammonium bromides, which have recently attracted much interest from the chemical and related industries. All the compounds tested demonstrated a dose-dependent inhibition of AMP deaminase activity. IC(50) ranged from 0.3µM for polycyclic musks to 500µM for N-glucopyranosyl trimethylammonium bromide. Analysis of Dixon plots showed the inhibition type for all the compounds to be noncompetitive. The results support the choice of such an assay for the prospective risk assessment of these compounds.

Evaluating the cytotoxicity of ionic liquids using human cell line HeLa.

This paper presents cytotoxicity data of selected imidazolium ionic liquids evaluated in vitro on the human tumor cell line HeLa. It was found that for 1-n-butyl-3-methylimidazolium entities the toxicity depends strongly on the associated anion; EC50 values are lowest for tetrafluoroborate. No direct dependence of the reduced effect concentration was found on elongating the short, methyl chain to ethyl or n-hexyl. Only for the ionic liquid with an n-decyl chain, the longest one studied, did higher hydrophobicity result in a EC50 one order of magnitude lower than that obtained with the n-butyl entity. The effect concentrations of imidazolium ionic liquids in the HeLa system used are lower than the values obtained for conventional organic solvents such as dichloromethane, toluene or xylene.

Modulation of G(2) arrest enhances cell death induced by the antitumor 1-nitroacridine derivative, Nitracrine.

Nitracrine (Ledakrin) is an antitumor drug which is activated by cellular enzymes and binds covalently to DNA. Previous studies have shown that covalent binding and crosslinking of DNA is associated with the cytotoxic and antitumor activities of this compound. In this study, cell cycle perturbations, effects on DNA synthesis and the cell death process initiated by Nitracrine were studied in murine leukemia L1210 cells. We show that exposure of L1210 cells to Nitracrine at the IC(99) concentration delayed progression through the S phase and transiently arrested cells in G(2)/M as found by flow cytometry. Higher drug concentration (2 x IC(99)) inhibited cell cycle progression in the S phase and induced rapid cell death. Both studied concentrations of the drug produced different effects on DNA synthesis as determined by bromodeoxyuridine incorporation, with a delay in the S phase progression at EC(99) concentration and irreversible arrest in early S phase at the higher dose (2 x IC(99)). At both concentrations of Nitracrine cell death occurred preferentially in the S phase as revealed by the TUNEL assay. When cells treated with the drug for 4 hours were post-incubated in the presence of 1 mM caffeine this led to rapid cell death and suppression of the G(2) arrest. This was associated with a about 10-fold increase in the cytotoxicity of Nitracrine. Similar effects were observed for another DNA crosslinking agent, cis-platinum, and to a lesser extent, for DNA topoisomerase I inhibitor, camptothecin. Together, our studies show that suppression of G(2) arrest induced by Nitracrine greatly enhances its cytotoxicity toward L1210 cells.

Recruitment of cdc2 kinase by DNA topoisomerase II is coupled to chromatin remodeling.

Although initiation of chromosome condensation during early prophase is linked temporally to the appearance of the mitotic cdc2 kinase in the nucleus, it is not known what targets the kinase to the nucleus and how this is coupled to chromatin remodeling. We now report that cdc2 kinase forms stable molecular complexes with the nuclear enzyme DNA topoisomerase II, which is associated with marked stimulation of both DNA binding and catalytic activity of topoisomerase II, albeit in a phosphorylation-independent manner. The molecular interaction is required for recruitment of cdc2 kinase, as shown by incubation of purified enzymes with chicken erythrocyte nuclei, which have neither endogenous topoisomerase II nor cdc2 kinase. The physical association between the two enzymes alters the DNA/topoisomerase II interaction as shown by pulse-field electrophoresis after incubation of intact nuclei with the specific topoisomerase II inhibitor VM-26. Furthermore, the presence of both enzymes, but not either enzyme alone, is accompanied by extensive chromatin remodeling converting the interphase nuclei into precondensation chromosomes with striking resemblance to early prophase structures. Our results reveal a novel property of cyclin-dependent kinases and demonstrate that the recruitment of cdc2 kinase by topoisomerase II is coupled to chromatin remodeling.

Does the antitumor cyclopropylpyrroloindole antibiotic CC-1065 cross-link DNA in tumor cells?

We have found that a cyclopropylpyrroloindole antibiotic, compound CC-1065 (benzo[1,2-b:4,3-b']dipyrrole-3(2H)-carboxamide, 7-[[1, 6-dihydro-4-hydroxy-5-methoxy-7-[(4,5,8, 8a-tetrahydro-7-methyl-4-oxocyclopropan[c]pyrrolo[3, 2-e]indol-2(1H)-yl)carbonyl]benzo[1,2-b:4, 3-b']dipyrrol-3(2H)-yl]-carbonyl]-1,6-dihydro-4-hydroxy-5-methoxy, (7bR,8aS)), forms interstrand DNA cross-links of an apparently covalent nature in HeLa S(3) cells. This compound induced interstrand cross-links at concentrations ranging from 0.1 to 1 nM/3 hr in whole cells, but these cross-links were absent or marginally low when the drug was added to cell lysates with inactivated cellular enzymes or isolated nuclei, which suggests that metabolic activation of the drug is a necessary step for DNA cross-linking to occur. In contrast, an analog of CC-1065, Bizelesin, which forms DNA-DNA cross-links by direct alkylation, induced interstrand DNA cross-links in both whole cells and in cell lysates. Interestingly, a demethoxy analog of CC-1065, Adozelesin, did not induce DNA cross-links under the same conditions. CC-1065 was found to be extremely potent in terms of concentrations required to cross-link DNA of tumor cells, and this may be related to its remarkable cytotoxic activity.

Distinct roles for recombinant cytosolic 5'-nucleotidase-I and -II in AMP and IMP catabolism in COS-7 and H9c2 rat myoblast cell lines.

Catabolism of AMP during ATP breakdown produces adenosine, which restores energy balance. Catabolism of IMP may be a key step regulating purine nucleotide pools. Two, cloned cytosolic 5'-nucleotidases (cN-I and cN-II) have been implicated in AMP and IMP breakdown. To evaluate their roles directly, we expressed recombinant pigeon cN-I or human cN-II at similar activities in COS-7 or H9c2 cells. During rapid (more than 90% in 10 min) or slower (30-40% in 10 min) ATP catabolism, cN-I-transfected COS-7 and H9c2 cells produced significantly more adenosine than cN-II-transfected cells, which were similar to control-transfected cells. Inosine and hypoxanthine concentrations increased only during slower ATP catabolism. In COS-7 cells, 5'-nucleotidase activity was not rate-limiting for inosine and hypoxanthine production, which was therefore unaffected by cN-II- and actually reduced by cN-I- overexpression. In H9c2 cells, in which 5'-nucleotidase activity was rate-limiting, only cN-II overexpression accelerated inosine and hypoxanthine formation. Guanosine formation from GMP was also increased by cN-II. Our results imply distinct roles for cN-I and cN-II. Under the conditions tested in these cells, only cN-I plays a significant role in AMP breakdown to adenosine, whereas only cN-II breaks down IMP to inosine and GMP to guanosine.

Resistance mechanisms associated with altered intracellular distribution of anticancer agents.

The resistance of tumor cells to anticancer agents remains a major cause of treatment failure in cancer patients. The term multidrug resistance (MDR) is used to define a resistance phenotype where cells are resistant to multiple drugs with no obvious structural resemblance and with different molecular targets. It is now clear that MDR is always multifactorial. The intracellular drug distribution is modified in many MDR cell lines, leading to increased drug sequestration in acidic vesicles, such as the trans-Golgi apparatus, recycling endosomes, and lysosomes, followed by transport to the plasma membrane and extrusion into the external medium. Since most anticancer agents target DNA or nuclear enzymes, sequestration of drug in cytoplasmic organelles will lead to decreased drug-target interaction and thereby, decreased cytotoxicity. Altered intracellular drug distribution is usually associated with the expression of drug efflux pumps, such as the P-glycoprotein and the multidrug resistance protein. Another common modification in MDR cells is alkalization of the intracellular pH. The relationship between these different resistance mechanisms is reviewed and a model proposed that suggests why these different resistance mechanisms are co-expressed in multiple cell lines.

Mitoxantrone and ametantrone induce interstrand cross-links in DNA of tumour cells.

We show here that mitoxantrone and ametantrone induce interstrand DNA cross-links in HeLa S3 cells. These cross-links were observed only in cellular system suggesting that metabolism of the drugs is a necessary step leading to DNA cross-linking. Biologically inactive analogue of mitoxantrone, compound NSC 321458, did not induce cross-links in DNA of tumour cells which suggests that DNA cross-linking is associated with the cytotoxic and anti-tumour activity of these compounds.

The interaction between p53 and DNA topoisomerase I is regulated differently in cells with wild-type and mutant p53.

DNA topoisomerase I is a nuclear enzyme involved in transcription, recombination, and DNA damage recognition. Previous studies have shown that topoisomerase I interacts directly with the tumor-suppressor protein p53. p53 is a transcription factor that activates certain genes through binding to specific DNA sequences. We now report that topoisomerase I can be stimulated by both latent and activated wild-type p53 as well as by several mutant and truncated p53 proteins in vitro, indicating that sequence-specific DNA-binding and stimulation of topoisomerase I are distinct properties of p53. These assays also suggest that the binding site for topoisomerase I on p53 is between amino acids 302 and 321. In living cells, the interaction between p53 and topoisomerase I is strongly dependent on p53 status. In MCF-7 cells, which have wild-type p53, the association between the two proteins is tightly regulated in a spatial and temporal manner and takes place only during brief periods of genotoxic stress. In marked contrast, the two proteins are constitutively associated in HT-29 cells, which have mutant p53. These findings have important implications for both cellular stress response and genomic stability, given the ability of topoisomerase I to recognize DNA lesions as well as to cause illegitimate recombination.

The mechanism of adenosine formation in cells. Cloning of cytosolic 5'-nucleotidase-I.

Adenosine increases blood flow and decreases excitatory nerve firing. In the heart, it reduces rate and force of contraction and preconditions the heart against injury by prolonged ischemia. Based on indirect kinetic arguments, an AMP-selective cytosolic 5'-nucleotidase designated cN-I has been implicated in adenosine formation during ATP breakdown. The molecular identity of cN-I is unknown, although an IMP/GMP-selective cytosolic 5'-nucleotidase (cN-II) and an ecto-5'-nucleotidase (e-N) have been cloned. We utilized the high abundance of cN-I in pigeon heart to purify a 40-kDa subunit for partial protein sequencing and subsequent cDNA cloning. We obtained a full-length clone encoding a novel 40-kDa peptide, unrelated to cN-II or e-N, that was most abundant in heart, brain, and breast muscle. Immunolocalization in heart showed a striated cytoplasmic location, suggesting association with contractile elements. Transient expression in COS-7 cells, generated a 5'-nucleotidase that catalyzed adenosine formation from AMP, which was increased during ATP catabolism. In conclusion, the cloning and expression of cN-I provides definitive evidence of its ability to produce adenosine during ATP breakdown.

Alteration of the daunorubicin-triggered sphingomyelin-ceramide pathway and apoptosis in MDR cells: influence of drug transport abnormalities.

We have previously shown that in myeloid leukemic cells, daunorubicin (DNR) induces apoptosis via the activation of the sphingomyelin-ceramide pathway. We have now investigated sphingomyelin (SM) hydrolysis, ceramide generation, and apoptosis in vincristine-selected multidrug resistant (MDR) HL-60 cells (HL-60/Vinc), compared with their parental counterparts. We show that DNR triggers the SM cycle (stimulation of neutral sphingomyelinase, SM hydrolysis, and ceramide generation) and apoptosis in both parental and MDR cells, when used at isotoxic doses (ie., 1 and 100 microM for HL-60 and HL-60/Vinc, respectively). However, in MDR cells treated with either 10 microM DNR or 1 microM DNR in association with the P-glycoprotein (P-gp) blocker verapamil (treatment conditions which yield an intracellular DNR concentration similar to that achieved with 1 microM in the parental cells), we were unable to detect SM hydrolysis, ceramide generation and apoptosis. This implies that inhibition of the DNR-induced SM cycle in MDR cells is not directly related to P-gp. We have also investigated the influence of intracellular drug localization on the DNR-induced SM-cycle in HL-60/Vinc cells. In these cells, DNR at 10 microM is mainly localized in cytoplasmic vesicles, while the drug is diffusely distributed when used at 100 microM. A diffuse distribution pattern was also observed when MDR cells were treated with 1 microM DNR in association with the cyclosporine derivative PSC-833, but not with verapamil. In parallel, PSC-833, but not verapamil, restored the induction of the SM cycle and the apoptotic potential of DNR, and markedly increased drug cytotoxicity in MDR cells. Our results suggest that altered intracellular drug transport plays an important role in limiting ceramide generation and cell death in MDR cells.

Dual mechanism of daunorubicin-induced cell death in both sensitive and MDR-resistant HL-60 cells.

Exposure of some acute myeloid leukaemia (AML) cells to daunorubicin leads to rapid cell death, whereas other AML cells show natural drug resistance. This has been attributed to expression of functional P-glycoprotein resulting in reduced drug accumulation. However, it has also been proposed that P-glycoprotein-expressing multidrug-resistant (MDR) cells are inherently defective for apoptosis. To distinguish between these different possibilities, we have compared the cell death process in a human AML cell line (HL-60) with a MDR subline (HL-60/Vinc) at doses that yield either similar intracellular daunorubicin concentrations or comparable cytotoxicity. Adjustment of the dose to obtain the same intracellular drug accumulation in the two cell lines did not result in equal cytotoxicity, suggesting the presence of additional resistance mechanisms in the P-glycoprotein-expressing HL-60/Vinc cells. However, at equitoxic doses, similar cell death pathways were observed. In HL-60 cells, daunorubicin induced rapid apoptosis at 0.5-1 microM and delayed mitotic cell death at 0.1 microM. These concentrations are within the clinical dose range. Similarly, HL-60/Vinc cells underwent apoptosis at 50-100 microM daunorubicin and mitotic cell death at 10 microM. These results show, for the first time, that anthracyclines can induce cell death by a dual mechanism in both sensitive and MDR cells. Our results also show that not only the cytotoxicity, but also the kinetics and mechanism of cell death, are dose dependent. Interestingly, regrowth was observed only in association with delayed cell death and the formation of enlarged, often polyploid, cells with micronucleation, suggesting that morphological criteria may be useful to evaluate treatment efficacy in patients with myeloid leukaemias.

Activity of IMP- and AMP-preferring isoforms of 5'-nucleotidase from human seminal plasma with AMP analogues.

AMP analogues modified at various positions of the molecule were checked as substrates for the two soluble isoforms of 5'-nucleotidase from human seminal plasma. These isoforms were isolated to near homogeneity by affinity chromatographies. AMP derivatives were differently dephosphorylated by both the isoforms depending on the site of modification in the natural compound. Changes in the phosphate moiety reduced significantly hydrolysis by the IMP-preferring form, whereas the AMP-preferring form was less affected. The AMP-preferring form was characterized by a relatively broad specificity toward substrate analogues indicating that the binding domains for the phosphate moiety of these isoforms are not identical. Substitutions at the C-8 adenine base reduced the hydrolysis rate of both the enzymes and variations of the syn-anti conformational equilibrium resulted in different effects on catalysis by both forms. Therefore, the orientation of the heterocyclic base around the glycosidic bond may not be the crucial factor affecting binding and catalytic activity. Hydrogen bonding potential of base N-7 was essential for the binding and catalysis of the IMP- but not of the AMP-preferring form. This was the most striking difference between the studied isoforms. Modifications and substitutions of 6-amino function, better accepted by the IMP-preferring form than by the AMP-preferring form, indicated that no essential hydrogen bonding is required for catalytic activity. The binding was however significantly slowed in 6-SH-PuMP. Hydrogen bonding potential of N-1 was significant for the hydrolysis rate of the IMP- but not of the AMP-preferring form. We suggest that these human seminal plasma isoforms of soluble 5'-nucleotidase, characterized by unique features, may represent the tissue-specific expression of the polymorphic gene.

DNA topoisomerases as repair enzymes: mechanism(s) of action and regulation by p53.

DNA topoisomerases regulate the organization of DNA and are important targets for many clinically used antineoplastic agents. In addition, DNA topoisomerases modulate the cellular sensitivity toward a number of DNA damaging agents. Increased topoisomerase II activities were shown to contribute to the resistance of both nitrogen mustard- and cisplatin-resistant cells. Similarly, cells with decreased topoisomerase II levels show increased sensitivity to cisplatin, carmustine, mitomycin C and nitrogen mustard. Recent studies propose that topoisomerases may be involved in damage recognition and DNA repair at several different levels including: 1) the initial recognition of DNA lesions; 2) DNA recombination; and 3) regulation of DNA structure. The stress-activated oncogene suppressor protein p53 can modulate the activity of at least three different human topoisomerases, either directly by molecular associations or by transcriptional regulation. Since DNA topoisomerases have considerable recombinase activities, inappropriately activated topoisomerases in tumor cells lacking functional p53 may contribute to the genetic instability of these cells.